Skip to main content
Glama
Gonzih

nexus-convergence-mcp

by Gonzih

list_model_disagreements

Reveals direct contradictions, disputed claims, and low-similarity pairs between model responses on a query, with similarity scores. Use it to analyze divergence and identify areas of disagreement.

Instructions

Return where models diverged on a query — direct contradictions (inversions), disputed claims, and low-similarity pairs with their similarity scores. Use this to understand the friction space: inversions are not errors, they are intelligence.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
query_idYesThe run/query ID returned by converge_query

Implementation Reference

  • The `listModelDisagreements` handler function. Fetches a run by queryId from the convergence service, retrieves its consensus analysis, and returns disagreements data including inversions, disputed claims, similarity matrix, and low-similarity pairs.
    export async function listModelDisagreements(queryId: string): Promise<{
      query_id: string;
      total_disagreements: number;
      inversions: ConsensusAnalysis['inversion_pairs'];
      disputed_claims: string[];
      similarity_matrix: ConsensusAnalysis['similarity_matrix'];
      low_similarity_pairs: ConsensusAnalysis['similarity_matrix'];
    }> {
      // Get run from convergence service to find analysis ID
      const run = await get<{ consensus_analysis_id?: string; model_responses?: unknown[] }>(
        `${CONVERGENCE_SERVICE_URL}/runs/${queryId}`
      );
    
      if (!run.consensus_analysis_id) {
        return {
          query_id: queryId,
          total_disagreements: 0,
          inversions: [],
          disputed_claims: [],
          similarity_matrix: [],
          low_similarity_pairs: [],
        };
      }
    
      const analysis = await getConsensusAnalysis(run.consensus_analysis_id);
    
      const lowSimilarityPairs = analysis.similarity_matrix.filter((p) => p.score < 0.5);
    
      return {
        query_id: queryId,
        total_disagreements: analysis.inversion_pairs.length + analysis.disputed_claims.length,
        inversions: analysis.inversion_pairs,
        disputed_claims: analysis.disputed_claims,
        similarity_matrix: analysis.similarity_matrix,
        low_similarity_pairs: lowSimilarityPairs,
      };
    }
  • src/index.ts:127-142 (registration)
    Tool registration: defines the tool name 'list_model_disagreements', its description, and input JSON Schema (requires 'query_id' string). Registered in the ListToolsRequestSchema handler.
    {
      name: 'list_model_disagreements',
      description:
        'Return where models diverged on a query — direct contradictions (inversions), disputed claims, and low-similarity pairs with their similarity scores. ' +
        'Use this to understand the friction space: inversions are not errors, they are intelligence.',
      inputSchema: {
        type: 'object' as const,
        required: ['query_id'],
        properties: {
          query_id: {
            type: 'string',
            description: 'The run/query ID returned by converge_query',
          },
        },
      },
    },
  • Tool call dispatch: Zod schema validation (query_id string) and call to the handler function. Part of the CallToolRequestSchema switch-case.
    case 'list_model_disagreements': {
      const schema = z.object({ query_id: z.string() });
      const params = schema.parse(args);
      const result = await listModelDisagreements(params.query_id);
    
      return {
        content: [
          {
            type: 'text' as const,
            text: JSON.stringify(result, null, 2),
          },
        ],
      };
  • Helper function `getConsensusAnalysis` used by `listModelDisagreements` to fetch consensus data from the consensus service.
    export async function getConsensusAnalysis(analysisId: string): Promise<ConsensusAnalysis> {
      return get<ConsensusAnalysis>(`${CONSENSUS_SERVICE_URL}/analysis/${analysisId}`);
  • Generic HTTP GET helper used to fetch data from the convergence and consensus services.
    async function get<T>(url: string): Promise<T> {
      const response = await fetch(url, { headers: authHeaders() });
    
      if (!response.ok) {
        const text = await response.text();
        throw new Error(`HTTP ${response.status}: ${text}`);
      }
    
      return response.json() as Promise<T>;
    }
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided; description does not mention authentication, rate limits, or side effects. It adds the behavioral nuance that inversions are not errors, but overall minimal disclosure for a tool with no annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, no unnecessary words, front-loaded with core function and then usage guidance. Every sentence adds value.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with one parameter and no output schema, the description adequately explains what is returned (inversions, disputed claims, low-similarity pairs with scores) and why to use it. No gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema covers 100% of the single parameter (query_id) with a complete description. The tool description adds no additional parameter meaning beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clearly states the tool returns where models diverged, specifying types of disagreements (inversions, disputed claims, low-similarity pairs) and similarity scores. Distinct from siblings (check_compliance, converge_query, get_evidence_ladder).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly says to use it to understand the friction space, including the insight that inversions are not errors. Lacks explicit alternatives or when-not-to-use, but provides clear usage context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Gonzih/nexus-convergence-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server